submission_id: hastagaras-halu-8b-llama_2564_v8
developer_uid: Hastagaras
status: inactive
model_repo: Hastagaras/Halu-8B-Llama3-v0.3
reward_repo: ChaiML/reward_gpt2_medium_preference_24m_e2
generation_params: {'temperature': 1.05, 'top_p': 0.95, 'min_p': 0.05, 'top_k': 100, 'presence_penalty': 0.0, 'frequency_penalty': 0.0, 'stopping_words': ['\n'], 'max_input_tokens': 512, 'best_of': 16, 'max_output_tokens': 64}
formatter: {'memory_template': "<|start_header_id|>system<|end_header_id|>\n\n{bot_name}'s Persona: {memory}\n\n", 'prompt_template': '{prompt}<|eot_id|>', 'bot_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}: {message}<|eot_id|>', 'user_template': '<|start_header_id|>user<|end_header_id|>\n\n{user_name}: {message}<|eot_id|>', 'response_template': '<|start_header_id|>assistant<|end_header_id|>\n\n{bot_name}:', 'truncate_by_message': False}
reward_formatter: {'memory_template': "{bot_name}'s Persona: {memory}\n####\n", 'prompt_template': '{prompt}\n<START>\n', 'bot_template': '{bot_name}: {message}\n', 'user_template': '{user_name}: {message}\n', 'response_template': '{bot_name}:', 'truncate_by_message': False}
timestamp: 2024-05-26T04:21:02+00:00
model_name: morebattleplease
model_eval_status: success
model_group: Hastagaras/Halu-8B-Llama
num_battles: 10056
num_wins: 5399
celo_rating: 1202.72
propriety_score: 0.0
propriety_total_count: 0.0
submission_type: basic
model_architecture: LlamaForCausalLM
model_num_parameters: 8030261248.0
best_of: 16
max_input_tokens: 512
max_output_tokens: 64
display_name: morebattleplease
ineligible_reason: propriety_total_count < 5000
language_model: Hastagaras/Halu-8B-Llama3-v0.3
model_size: 8B
reward_model: ChaiML/reward_gpt2_medium_preference_24m_e2
us_pacific_date: 2024-05-25
win_ratio: 0.5368933969769292
Resubmit model
Running pipeline stage MKMLizer
Starting job with name hastagaras-halu-8b-llama-2564-v8-mkmlizer
Waiting for job on hastagaras-halu-8b-llama-2564-v8-mkmlizer to finish
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ╔═════════════════════════════════════════════════════════════════════╗
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ _____ __ __ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ / _/ /_ ___ __/ / ___ ___ / / ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ / _/ / // / |/|/ / _ \/ -_) -_) / ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ /_//_/\_, /|__,__/_//_/\__/\__/_/ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ /___/ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ Version: 0.8.14 ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ Copyright 2023 MK ONE TECHNOLOGIES Inc. ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ https://mk1.ai ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ The license key for the current software has been verified as ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ belonging to: ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ Chai Research Corp. ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ Account ID: 7997a29f-0ceb-4cc7-9adf-840c57b4ae6f ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ Expiration: 2024-07-15 23:59:59 ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ║ ║
hastagaras-halu-8b-llama-2564-v8-mkmlizer: ╚═════════════════════════════════════════════════════════════════════╝
hastagaras-halu-8b-llama-2564-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/huggingface_hub/utils/_deprecation.py:131: FutureWarning: 'list_files_info' (from 'huggingface_hub.hf_api') is deprecated and will be removed from version '0.23'. Use `list_repo_tree` and `get_paths_info` instead.
hastagaras-halu-8b-llama-2564-v8-mkmlizer: warnings.warn(warning_message, FutureWarning)
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Downloaded to shared memory in 15.125s
hastagaras-halu-8b-llama-2564-v8-mkmlizer: quantizing model to /dev/shm/model_cache
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Saving flywheel model at /dev/shm/model_cache
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Loading 0: 0%| | 0/291 [00:00<?, ?it/s] Loading 0: 4%|▍ | 13/291 [00:00<00:02, 128.42it/s] Loading 0: 11%|█ | 31/291 [00:00<00:01, 148.35it/s] Loading 0: 17%|█▋ | 49/291 [00:00<00:01, 154.85it/s] Loading 0: 23%|██▎ | 67/291 [00:00<00:01, 156.87it/s] Loading 0: 29%|██▊ | 83/291 [00:00<00:02, 81.19it/s] Loading 0: 33%|███▎ | 96/291 [00:00<00:02, 90.82it/s] Loading 0: 39%|███▉ | 113/291 [00:01<00:01, 105.63it/s] Loading 0: 45%|████▌ | 131/291 [00:01<00:01, 119.13it/s] Loading 0: 51%|█████ | 149/291 [00:01<00:01, 129.65it/s] Loading 0: 57%|█████▋ | 167/291 [00:01<00:00, 139.22it/s] Loading 0: 64%|██████▍ | 187/291 [00:01<00:01, 94.42it/s] Loading 0: 69%|██████▉ | 201/291 [00:01<00:00, 102.74it/s] Loading 0: 75%|███████▌ | 219/291 [00:01<00:00, 116.38it/s] Loading 0: 81%|████████▏ | 237/291 [00:02<00:00, 127.16it/s] Loading 0: 88%|████████▊ | 255/291 [00:02<00:00, 136.29it/s] Loading 0: 94%|█████████▍| 273/291 [00:02<00:00, 142.52it/s] Loading 0: 99%|█████████▉| 289/291 [00:07<00:00, 9.93it/s] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
hastagaras-halu-8b-llama-2564-v8-mkmlizer: quantized model in 23.028s
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Processed model Hastagaras/Halu-8B-Llama3-v0.3 in 40.731s
hastagaras-halu-8b-llama-2564-v8-mkmlizer: creating bucket guanaco-mkml-models
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Bucket 's3://guanaco-mkml-models/' created
hastagaras-halu-8b-llama-2564-v8-mkmlizer: uploading /dev/shm/model_cache to s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /dev/shm/model_cache/config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8/config.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /dev/shm/model_cache/special_tokens_map.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8/special_tokens_map.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer_config.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8/tokenizer_config.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /dev/shm/model_cache/tokenizer.json s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8/tokenizer.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /dev/shm/model_cache/flywheel_model.0.safetensors s3://guanaco-mkml-models/hastagaras-halu-8b-llama-2564-v8/flywheel_model.0.safetensors
hastagaras-halu-8b-llama-2564-v8-mkmlizer: loading reward model from ChaiML/reward_gpt2_medium_preference_24m_e2
hastagaras-halu-8b-llama-2564-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py:913: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
hastagaras-halu-8b-llama-2564-v8-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-2564-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py:757: FutureWarning: The `use_auth_token` argument is deprecated and will be removed in v5 of Transformers. Please use `token` instead.
hastagaras-halu-8b-llama-2564-v8-mkmlizer: warnings.warn(
hastagaras-halu-8b-llama-2564-v8-mkmlizer: /opt/conda/lib/python3.10/site-packages/torch/_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
hastagaras-halu-8b-llama-2564-v8-mkmlizer: return self.fget.__get__(instance, owner)()
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Saving model to /tmp/reward_cache/reward.tensors
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Saving duration: 0.401s
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Processed model ChaiML/reward_gpt2_medium_preference_24m_e2 in 7.297s
hastagaras-halu-8b-llama-2564-v8-mkmlizer: creating bucket guanaco-reward-models
hastagaras-halu-8b-llama-2564-v8-mkmlizer: Bucket 's3://guanaco-reward-models/' created
hastagaras-halu-8b-llama-2564-v8-mkmlizer: uploading /tmp/reward_cache to s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/special_tokens_map.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/special_tokens_map.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/merges.txt s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/merges.txt
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/vocab.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/vocab.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/config.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/tokenizer_config.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/tokenizer_config.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/tokenizer.json s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/tokenizer.json
hastagaras-halu-8b-llama-2564-v8-mkmlizer: cp /tmp/reward_cache/reward.tensors s3://guanaco-reward-models/hastagaras-halu-8b-llama-2564-v8_reward/reward.tensors
Job hastagaras-halu-8b-llama-2564-v8-mkmlizer completed after 62.58s with status: succeeded
Stopping job with name hastagaras-halu-8b-llama-2564-v8-mkmlizer
Pipeline stage MKMLizer completed in 63.41s
Running pipeline stage MKMLKubeTemplater
Pipeline stage MKMLKubeTemplater completed in 0.08s
Running pipeline stage ISVCDeployer
Creating inference service hastagaras-halu-8b-llama-2564-v8
Waiting for inference service hastagaras-halu-8b-llama-2564-v8 to be ready
Inference service hastagaras-halu-8b-llama-2564-v8 ready after 60.74928641319275s
Pipeline stage ISVCDeployer completed in 66.74s
Running pipeline stage StressChecker
Received healthy response to inference request in 2.985084056854248s
Received healthy response to inference request in 1.4267215728759766s
Received healthy response to inference request in 1.4015743732452393s
Received healthy response to inference request in 1.3598196506500244s
Received healthy response to inference request in 1.31142258644104s
5 requests
0 failed requests
5th percentile: 1.321101999282837
10th percentile: 1.3307814121246337
20th percentile: 1.3501402378082275
30th percentile: 1.3681705951690675
40th percentile: 1.3848724842071534
50th percentile: 1.4015743732452393
60th percentile: 1.4116332530975342
70th percentile: 1.421692132949829
80th percentile: 1.7383940696716311
90th percentile: 2.36173906326294
95th percentile: 2.6734115600585935
99th percentile: 2.9227495574951172
mean time: 1.6969244480133057
Pipeline stage StressChecker completed in 9.10s
Running pipeline stage DaemonicModelEvalScorer
Pipeline stage DaemonicModelEvalScorer completed in 0.03s
Running pipeline stage DaemonicSafetyScorer
Pipeline stage DaemonicSafetyScorer completed in 0.03s
Running M-Eval for topic stay_in_character
hastagaras-halu-8b-llama_2564_v8 status is now deployed due to DeploymentManager action
M-Eval Dataset for topic stay_in_character is loaded
hastagaras-halu-8b-llama_2564_v8 status is now inactive due to auto deactivation removed underperforming models

Usage Metrics

Latency Metrics